1. Field of the Invention
The present invention relates to a semiconductor memory card that stores audio data and control data, and to a playback apparatus, recording apparatus, playback method, recording method, and computer-readable recording medium relating to such a semiconductor memory card. In particular, the present invention relates to improved storage of management information and audio data distributed as content by a content distribution service, such as an electronic music distribution service.
2. Description of the Background Art
Recent years have witnessed the gradual introduction of the hardware infrastructure necessary for the electronic distribution of music. This gives rise to the potential for great change in the music industry, where products have been conventionally distributed as packaged software using media such as compact discs (CDs) and cassette tapes.
Electronic music contents (i.e., songs and albums) can be delivered to consumers by having the consumer's personal computer download contents from a server computer operated by a record label. To listen to the downloaded digital music on a portable player, the user needs to store the music data onto a portable recording medium. At present, the most suitable media for storing electronically distributed music data are semiconductor memory cards.
As examples of such, flash ATA cards and COMPACT FLASH cards are already available. Such semiconductor memory cards include a semiconductor device called flash memory (EEPROM—Electrically Erasable Programmable Read-Only Memory). Flash memory is capable of data reads and writes at much higher speeds than MD (MiniDisc) or CD-R (Compact Disc-Recordable). This means that digital music can be transferred in a short time, in spite of its large data size.
As a major disadvantage, semiconductor memory cards carry the risk of allowing users to make illegal copies of copyrighted music that has been downloaded from an electronic music distribution service. Since semiconductor memory cards allow data to be written at higher speeds than CD-R or MD, copying is thought to be a more serious problem for such memory cards. In order to overcome the potential dangers regarding copyright infringement, digital music has to be encrypted using a secure encryption method before being stored in a semiconductor memory card.
One storage method that takes into account the need to prevent unauthorized copying is the title storage method used under DVD-Audio standard. As one example of this method, a “title”, which corresponds to a conventional music album, includes a plurality of “contents”, which correspond to tracks on the album. The contents that compose a title are encrypted using an encryption key, called the “title key”, chosen by the disc producer before being recorded on a DVD-Audio disc. This title key is encrypted using an encryption key (usually called the “disc key”) that is unique to each DVD-Audio disc and is stored in a sector header region of a DVD-Audio disc. This disc key is itself encrypted using an encryption key (usually called the “master key”) chosen by the manufacturers of content decoding apparatuses and is recorded in the lead-in region of the DVD-Audio disc. The sector header region and lead-in region cannot be accessed by ordinary users, making it extremely difficult for users to illegally obtain the title key recorded on a DVD-Audio disc.
In comparison to magnetic or optical storage media, semiconductor memory cards have a limited storage capacity, so that it is normally necessary to compress digital music with a high compression ratio when storing it onto a semiconductor memory card. One encoding method for achieving a sufficiently high compression ratio for digital music is MPEG2-AAC (Motion Pictures Experts Group 2-Advanced Audio Coding). One characteristic of MPEG2-AAC compression is that it makes use of the limitations of human hearing and so changes the bit length of the data assigned to each audio frame, an audio frame being the smallest playback unit and representing around 20 ms of audio. Data with longer bit lengths is assigned to audio frames that have many frequencies within the range of human hearing, while the shorter bit lengths are assigned to audio frames with fewer of such sounds or frequencies outside the range of human hearing.
Since the amount of data assigned to each audio frame in MPEG2-AAC depends on the number of audible frequencies in the frame (or in other words, because MPEG2-AAC uses variable-bitrate (VBR) encoding), high-quality audio contents can be obtained even at high rates of compression. Such audio contents are suited to distribution on a public network and to storage onto semiconductor memory cards that have a limited storage capacity.
First Problem
When contents are stored according to conventional methods, decoding the title key used to encrypt the music contents will enable the user to decrypt all of the music contents recorded on a recording medium. This gives rise to the first problem of the exposure of a single title key making it easy for users to decrypt all of the tracks stored on a semiconductor memory card.
While title keys will seldom be exposed, such exposure will result in an immeasurable loss to the copyright holder. With the great advancements in the processing power of home computers in recent years, it is becoming increasingly difficult to say that a title key used to encrypt digital music will completely safe from decoding. This gives rise to demands for a data construction that will minimize the damage to copyright holders when a title key is exposed.
Second Problem
As copyright protection is necessary for digital music that is to be distributed by electronic music distribution, such music is usually distributed in an encrypted form. Encryption is also required for digital music stored in a semiconductor memory card. However, this gives rise to a second problem that a user who has paid the proper price to purchase digital music will not be able to freely edit the music when it is stored in an encrypted manner on a semiconductor memory card. If the music contents are stored in an encrypted form, it will be very difficult for the user to change the order of tracks or to partially delete tracks. Considering that the user has paid the proper price, it is not desirable to restrict his/her ability to edit music contents in this way.
MiniDisc (MD) recorders, which can be used for recording music in the same way as a semiconductor memory card, allow a variety of track editing functions through the provision of a TOC (Table of Contents). Such functions include the rearranging of the playback order of tracks, the division of tracks, and the combining of tracks into a single track. If semiconductor memory card recorders are unable to provide the same functions as conventional MD recorders, it is believed that consumers will regard semiconductor memory card players as inferior to MD recorders, thereby damaging the commercial potential of semiconductor memory card products.
Third Problem
To provide special playback functions for digital music that has been subjected to VBR encoding, as under MPEG2-AAC, playback apparatuses need to be equipped with large-capacity memories. This raises the manufacturing cost of such apparatuses, and poses a third problem for the background art.
The special playback functions provided by MD or CD players include the ability to start playback from any track on a disc (specifying the playback position), a music search function that plays back intermittent bursts of music to enable users to skip through tracks forwards or backwards at high speed, and a time search function whereby users can have the playback start from a position inputted as a time measured from the start of the disc. To capture the market currently held by MD or CD players, it is essential for playback apparatuses of semiconductor memory cards to provide the same special playback functions as MD players. When music contents are subjected to constant bitrate (CBR) encoding, playback from a position specified using a time code (such a point one or two minutes from the start of a track) can be performed simply by referring to an address that is offset by an integer multiple of the data size of the unit playback time. However, when music contents are encoded using a VBR method such as MPEG2-AAC, the positions corresponding to one or two minutes ahead of the current position will seldom be offset by an integer multiple of the data size of the unit playback time. As a result, a player will need to refer to a time search table produced in advance to show which addresses correspond to the points one minute and two minutes further ahead.
While a time search table for a short track will not need to include a large number of playback positions, this cannot be said for the time search tables of long tracks, so that the time search tables of long tracks are very large. To provide special playback features, a playback apparatus has to access the time search table having first loaded it into its memory. Since long tracks have large time search tables, this means that a playback apparatus has to be provided with a large memory for storing the time search table. This also increases the manufacturing costs of playback apparatuses.